deslop
Pass
Audited by Gen Agent Trust Hub on Aug 27, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes code diffs from external branches, which serves as an ingestion point for potentially untrusted data. If an attacker embeds malicious instructions within code comments or string literals in a branch, the agent might inadvertently execute them while performing 'deslopping' tasks.
- Ingestion points: SKILL.md instructs the agent to 'Check the diff against main'.
- Boundary markers: Absent. The instructions do not include delimiters or warnings to ignore commands embedded within the code being processed.
- Capability inventory: The skill is designed to modify code, implying file-write and file-read capabilities on the local filesystem.
- Sanitization: Absent.
Audit Metadata